Traditional telecommunications networks, especially wireless communications systems such as the Global System for Mobile communications (GSM) and Universal Mobile Telecommunications System (UMTS) have the merits of rich services and strong network control and therefore have been widely adopted. With the continuous development and improvement of network technologies, higher data rates are required. But, because of limited frequency resources and widely variable transport environments, wireless communications systems are unable to provide higher access rates. Therefore, wideband radio access becomes a key solution for higher access rates. Wideband radio access technologies represented by Wireless Local Area Network (WLAN) and World Interoperability for Microwave Access (WiMAX) can provide high rate wideband radio access services. They also support nomadic and mobile applications. The access capability of a wireless communications network is therefore much stronger. The convergence of the mobile communications network and wideband radio access technology becomes an evolutionary trend of telecommunications networks.
In a moving scenario of a wideband radio access network, service continuity is required when a User Equipment (UE) is handed over between different access networks. Because the network prefix varies with the access network link, when a UE is handed over from a source network link to a target access network link, the network prefix of the UE's Internet Protocol (IP) address in the target access network will be different from the network prefix of the IP address in the source network. As a result, in the moving scenario, routing based on a common IP network prefix is unable to forward packets to the target access network position of the UE. If the UE updates its IP address during a handover process, continuity of ongoing services will be impossible.
To solve the above issue, Mobility IP (MIP) is adopted so that a UE is able to keep its Home-of Address (HOA) unchanged when moving. The MIP technology is briefed below.
The basic principle of MIP is that one UE is associated with two IP addresses, namely, Home-of Address (HOA) and Care-of Address (COA), so that the UE is able to maintain its HOA when moving. The UE obtains a HOA from the home network. When the UE moves outside the home network, the UE obtains a COA of the current network from a foreign access link mobile proxy and notifies the home link mobile proxy of the COA. The home link mobile proxy binds the HOA and the COA of the UE and sets up a tunnel between the home link mobile proxy and the COA (foreign access link mobile proxy). Afterwards, the home link mobile proxy sends packets destined for the HOA of the UE to the COA of the UE via the tunnel so as to complete routing of the packets.
Existing telecommunications networks allow a UE to access one or more packet data services of a Packet Data Network (PDN). The multiple packet data services are identified by Access Point Names (APNs). APNs of packet data services the UE wants to access may be pre-configured in the access network or be provided by the UE for the access network. The access network establishes the connectivity from the UE to an appropriate Packet Data Network Gateway (PDN GW) according to an APN. The PDN GW then establishes the connectivity to the appropriate PDN according to the APN.
For efficient management and utilization of network resources, when the UE leaves the network, it is necessary to release resources allocated for the UE in time, including radio channels, bearers, various tunnels and storage, so as to increase the utilization of radio resources.
After network evolution, the UE may obtain required services through one access network of three evolved network structures; after the UE enters the network, if the UE wants to leave the current source network and enter a target network, it is necessary to hand over the UE between the three evolved networks. The bearer in the source network must be handed over to the target network. This includes a target network bearer setup process and a source network bearer release process. To guarantee the continuity of ongoing services when the UE is in handover, with respect to handover between the three evolved networks, the following requirements must be met:                least impact on the handovered networks;        least impact on the UE;        least coupling between handovered networks; and        assured continuity of services.        
To meet the above requirements and to offer access choices and service diversity, it is necessary to consider a more optimized handover method between evolved networks so that inter-RAT handover is completed quickly and accurately in real time.
FIG. 1 is a schematic drawing illustrating a structure of optimized handover between three evolved networks in the prior art. In FIG. 1, the interface between a home SGW and a PDN GW is S5 (not shown in the figure) and the interface between different evolved networks is S71, where communications are based on a tunneling protocol. With the S71 interface, the communications between the UE and the target network are transparent to the source network so as to reduce the impact of the handover on the source network and minimize the coupling between the source network and the target network.When implementing the present invention, however, the inventor finds that the prior art has at least the following defect:
In the prior art, specific processes involved in optimized handover are not decomposed. This means there is no complete mechanism or procedure for optimized handover between evolved networks. As a result, continuity of services can not be assured during a handover process.